Artificial intelligence can be used to better monitor Maine’s forests, finds UMaine study – UMaine News

Monitoring and measuring forest ecosystems is a complex challenge due to an existing combination of software, logging systems and computing environments that require increasing amounts of energy to operate. The University of Maine’s Wireless Sensor Networks (WiSe-Net) Laboratory has developed a novel method of using artificial intelligence and machine learning to make soil moisture monitoring more energy and cost efficient – one that could be used to improve the measurement in the vast forest more efficient ecosystems of Maine and beyond.

Soil moisture is an important variable in both forested and agricultural ecosystems, particularly under the recent drought conditions in Maine over the past few summers. Despite the robust soil moisture monitoring networks and large, freely available databases, the cost of commercial soil moisture sensors and the power consumption they require to operate can be prohibitive for researchers, foresters, farmers, and others who track the condition of the soil.

Together with researchers from the University of New Hampshire and the University of Vermont, UMaine’s WiSe-Net designed a wireless sensor network that uses artificial intelligence to learn how to be more energy efficient when monitoring soil moisture and processing the data. The research was funded by a National Science Foundation grant.

“AI can learn from the environment, predict wireless connection quality and incoming solar energy to use limited energy efficiently and keep a resilient, low-cost network running longer and more reliably,” says Ali Abedi, principal investigator on the latest study and professor of Electrical and Computer Engineering from the University of Maine.

The software learns over time how to best utilize available network resources, helping to produce energy-efficient systems at a lower cost for large-scale surveillance compared to existing industry standards.

WiSe-Net also worked with Aaron Weiskttel, director of the Center for Sustainable Forest Research, to ensure all hardware and software research is science-based and tailored to research needs.

“Soil moisture is a major driver of tree growth, but it changes rapidly, both daily and seasonally,” says Weiskttel. “We lacked the ability to monitor effectively on a large scale. In the past we used expensive sensors that collected at fixed intervals – say every minute – but were not very reliable. A cheaper and more robust sensor with wireless capabilities like this really opens the door to future applications for researchers and practitioners alike.”

That to learn was published in Springer’s International Journal of Wireless Information Networks on August 9, 2022.

Although the system designed by the researchers focuses on soil moisture, the same methodology could be extended to other types of sensors such as ambient temperature, snow depth and more, as well as scaling the networks with more sensor nodes.

“Real-time monitoring of different variables requires different sample rates and power levels. An AI agent can learn these and adjust the frequency of data collection and transmission accordingly, rather than collecting and sending every single data point, which is not as efficient,” says Abedi.

Contact: Sam Schipani, [email protected]